1891.] Results from Observations made at Lovedale S.A. 27 
same way on the same sheet, the values assigned being placed along- 
side those of the previous evening. 
The same is done on a third night, and also on a fourth. The 
thought will at once present itself, “ will not the recorded magnitudes 
on the different nights become inextricably mixed.” The first night 
I write down the figures with an ordinary black pencil, the second 
night with a blue pencil, the third night with black ink, and the 
fourth with red ink. When this has been done on four nights an 
exact fac-simile of the position of the stars is taken on a clean sheet 
of drawing paper by pressing the marks with the sharp point of a 
pencil or puncturing the paper with a pin. 
With this new sheet another set of values may be obtained as 
before. After four or five sheets are obtained the observer may take 
up another zone, returning to the first one at the end of every month. 
This method is an exhaustive, perhaps also exhausting one; yet if 
carefully and systematically carried out it would result in the discovery 
of all variables above the 9th magnitude. One objection to it is its 
tediousness. It will be in the memory of many that when a search 
was begun for the planet Neptune, the methods adopted by Prof.. 
Challis and Dr. Galle were quite different. Prof. Challis carefully 
charted night after night all the stars in the region where Prof. 
Adams said the planet would be found, intending when he had 
leisure and a large number of observations to compare them. 
This method was slow but sure. Neptune was in the observer’s 
net, and only time and opportunity were needed to sort it out from 
the thousand and one other stars that had been charted down. The 
guiding idea, at least if Bremiker’s charts had not been available in Dr. 
Galle’s search was to look for a star “with a disc.” His method 
was the more rapid, but it depended entirely upon the correctness 
of the data given. Prof. Pickering has suggested a somewhat similar 
method of search after short period variables. It has been noticed 
that almost all short period variables are white, that they usually 
lie in that portion of the Milky Way between XII and XVIII R.A., 
and that they generally have a small companion about 3’ of are 
distant and with a position angle of 70° reckoned from the south 
towards the east. It has been suggested, therefore, that observers 
should sweep along this region and when they come across any star 
fulfilling these conditions to watch it carefully for a time. While 
very short period variables are white, ordinary variables and especially 
long period variables are red. It has been asserted also that when 
